Prof. S. J. SJtand — Saturation in Petrography. 489 



statements cau be made, hence there is no correhition. Hatch ^ 

 effects a classification by employing the mineral constitution, as 

 expressed by felspars and felspathoids, and the silica-percentage as 

 abscissa and ordinate in a cross-classification. It is obvious that the 

 method does justice neither to chemistry nor to mineralogy, since all 

 the constituents are not regarded ; and furthermore it is easy to show- 

 that such a classification makes little advance upon a simple linear 

 classification by felspars alone or by silica-percentage alone, because 

 when the felspars are abundant they largely determine the silica- 

 percentage of the rock. This means that in many cases the two 

 co-ordinates are not independent of each otlier, hence the purpose of 

 a cross-classification is not attained. 



Winchell" has recently proposed a classification on three co- 

 ordinates. One of these is mineral constitution, and is chiefly 

 concerned with the kind of felspar present. The second is geological 

 occurrence. Tlie third co-ordinate is the alkaline, peralkaline, or 

 alkalcic (i.e. alkali-calcic) character of the rock as determined (by 

 inspection) from consideration of the relation between alkalies, 

 silica, lime, etc., in all the minerals present. Thus the peralkaline 

 group consists of rocks containing felspathoids ; the alkaline group 

 contains chiefly alkali-felspars and alkali-pyroxenes, etc. The third 

 co-ordinate therefore expresses in a rough way the ratio of alkalies to 

 silica, thus — 



Alkaline = high alkalies, high silica. 

 Peralkaline = high alkalies, moderate or low silica. 

 Alkalcic = low' alkalies, moderate or low silica. 



This classification consequently affords only a very incomplete and 

 purely qualitative expression of chemical composition. 



There are many other classifications which might be passed in 

 review, but it would be waste of ink, for it is matter of common 

 knowledge that, whatever their merits may be, no one of them has 

 ever succeeded in so correlating mineralogy with chemical composition 

 as to satisfy the two opposed schools of petrography. Personally, 

 I have renounced the attempt. 1 believe it to be impossible. But 

 that does not mean that, in my view, one class of data must be 

 neglected. I believe that both chemical and mineralogical 

 classifications are necessary in order to " give us the greatest 

 command over our knowledge already acquired, and lead most 

 directly to the acquisition of more". And instead of yearning for the 

 unattainable, I should like to see all petrographers trying to agree 

 upon one chemical classification and one mineralo<iical classification 

 to meet our present needs. They might not be final, but what of 

 that? They would at least give us a common language in which to 

 compare and record our observations. Apothecaries' weight, based 

 on a grain of corn and other primitive standards, served science well 

 until the metric system was introduced. Fahrenheit and Centigrade 

 thermometers registered the observations which led to the conception 



1 Text-book of Petrology, 1909, 1914. 

 ^ Journal of Geology, xxi, 1913. 



